Adobe at NAACL 2022

July 11, 2022

Tags: AI & Machine Learning, Conferences, Natural Language Processing

In this NAACL 2022 paper, we propose a new method to generate image captions that are specific and cover detailed aspects of the image that distinguish it from others.

The 2022 Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL 2022) is being held from July 10 to 15, 2022, in Seattle. NAACL is one of the top research conferences on natural language processing. In recent years, deep learning approaches have been prominently featured in the papers presented at this conference. 

Adobe has co-authored a total of eleven papers at the conference, four papers at the main conference, four papers in the Findings category, one workshop paper, one paper at the *SEM 2022 conference, and one paper at SemEval 2022 (both co-located with NAACL 2022). The research topics range from named entity extraction, event detection, temporal dependency, document navigation, image captioning, transcript generation, and many more.  

Findings papers, while not accepted for publication in the main conference, were assessed by the NAACL 2022 Program Committee to contain solid work with sufficient substance, quality, and novelty to be included in the conference proceedings. 

Many of the accepted papers are the result of student internships or other collaborations with university students and faculty. Check out the Adobe Research Careers page to learn more about internships and full-time career opportunities with our team.

NAACL 2022 conference – Adobe co-authored papers 

Main conference papers

DocTime: A Document-level Temporal Dependency Graph Parser  
Puneet Mathur, Vlad I Morariu, Verena Kaynig-Fittkau, Jiuxiang Gu, Franck Dernoncourt, Quan Hung Tran, Ani Nenkova, Dinesh Manocha, Rajiv Jain 

DynamicTOC: Persona-based Table of Contents for Consumption of Long Documents  
Himanshu Maheshwari, Nethraa Sivakumar, Shelly Jain, Tanvi Karandikar, Vinay Aggarwal, Navita Goyal, Sumit Shekhar 

Joint Extraction of Entities, Relations, and Events via Modeling Inter-Instance and Inter-Label Dependencies
Minh Van Nguyen, Bonan Min, Franck Dernoncourt, Thien Huu Nguyen 

MINION: a Large-Scale and Diverse Dataset for Multilingual Event Detection 
Amir Pouran Ben Veyseh, Minh Van Nguyen, Franck Dernoncourt, Thien Huu Nguyen

Findings papers 

BehancePR: A Punctuation Restoration Dataset for Livestreaming Video Transcript 
Viet Dac Lai, Amir Pouran Ben Veyseh, Franck Dernoncourt, Thien Huu Nguyen 

Event Detection for Suicide Understanding
Luis Fernando Guzman-Nateras, Viet Dac Lai, Amir Pouran Ben Veyseh, Franck Dernoncourt, Thien Huu Nguyen 

Fine-grained Image Captioning with CLIP Reward 
Jaemin Cho, Seunghyun Yoon, Ajinkya Kale, Franck Dernoncourt, Trung Bui, Mohit Bansal 

Multimodal Intent Discovery from Livestream Videos
Adyasha Maharana, Quan Hung Tran, Seunghyun Yoon, Franck Dernoncourt, Trung Bui, Walter Chang, Mohit Bansal 

NAACL 2022 Workshop

Generating Complement Data for Aspect Term Extraction with GPT-2
Amir Pouran Ben Veyseh, Franck Dernoncourt, Bonan Min, Thien Huu Nguyen
Presented at the DeepLo Workshop

Paper at *SEM 2022 – Joint Conference on Lexical and Computational Semantics (co-located with NAACL 2022)

What do Large Language Models Learn about Scripts? 
Abhilasha Sancheti, Rachel Rudinger                        

Paper at SemEval 2022 – International Workshop on Semantic Evaluation (co-located with NAACL 2022)

SemEval 2022 Task 12: Symlink – Linking Mathematical Symbols to their Descriptions 
Viet Lai, Amir Pouran Ben Veyseh, Franck Dernoncourt, Thien Nguyen

Workshop co-organizers

Workshop on Large-scale Pre-trained Language Models
Franck Dernoncourt

Shared Task on Linking mathematical symbols to their descriptions at SemEval 2022
Franck Dernoncourt

2nd HCI + NLP Workshop
Ani Nenkova

Related Posts